In this study, 144 infants who had neonatal hypoglycemia were analyzed for their neurodevelopment by the Gesell scoring method to investigate their gross motor, fine motor, adaptability (including the abilities of fine-motor coordination for objects and scenes, hand-eye coordination, problem solving, and application tools), language, and social skills at 2 years old. We found that long and repeated neonatal hypoglycemia, especially that lasting for more than 24 h, affected neurodevelopment and was associated with a high risk of poor adaptability. Indeed, studies in newborns with hypoglycemia by magnetic resonance imaging have shown that edema occurs in the posterior occipital and cortex region, with symmetrical changes [
25,
26]. The occipital and cortex regions are somatosensory and visual control areas [
27], which impact cognitive skills, adaptability, and visual skills. Under hypoglycemic conditions, the liver glycogen reserves are insufficient. Once the blood sugar level reaches the lowest point, the synthesis of lipids, proteins, DNA, and RNA is limited or delayed because not enough energy is lied, thus affecting brain cell metabolism and development and eventually leading to neuronal necrosis. A high level of glucose is required for the occipital region because there are more neurons and synapses in this region [
28]. If hypoglycemia is not able to be quickly corrected, irreversible brain damage in the posterior occipital and cortex regions will result.
Neonatal hypoglycemia is a common metabolic disorder during the neonatal period. Volpe has indicated that continuous, repeated hypoglycemia can cause brain damage [
29]. In addition, Filan et al. have found that neonatal hypoglycemia can injure the occipital brain, resulting in long-term disability, visual impairment, and epilepsy [
28]. In this study, at a corrected age of 2 years old, no significant difference was found in any assessment score of neurodevelopment (including gross motor, fine motor, adaptability, language, and social skills) between the infants who had neonatal hypoglycemia and controls. This result seems similar with that of Christopher et al., who found that neonatal hypoglycemia is not related to adverse neurodevelopment at 2 years old [
30]. However, since neonatal hypoglycemia occurred at different times in group A, we further divided this group into A1 (neonatal hypoglycemia within 2 h of birth), A2 (neonatal hypoglycemia at 2–24 h of birth), and A3 (neonatal hypoglycemia at more than 24 h of birth). Interestingly, the adaptability scores in subgroups A2 and A3 were significantly lower than that of the control group (73.9 ± 6.6 vs. 87.9 ± 11.2,
p = 0.0243; 71.5 ± 8.9 vs. 87.9 ± 11.2,
p = 0.0138, respectively). This finding indicated that temporary hypoglycemia (within 2 h) did not induce neurodevelopmental injury. However, long and repeated hypoglycemia decreased adaptability development. In a follow-up study at 4.5 years old, neonatal hypoglycemia was found to increase the risk of poor executive function, visual skills, and fine motor skills [
2], especially in infants with hypoglycemia at more than 24 h after birth. This result is different from our findings. The reason might be because of the different observation age and assessment methods. Thus, further follow-up after a longer time period and with different methods is necessary. The etiology of adverse neurodevelopment caused by neonatal hypoglycemia is unclear. Filan et al. believe that transient hyperinsulinism is an independent risk factor for neonatal hypoglycemia [
28]. Thus, an animal model study would be valuable. There is no clear consensus on the management of neonatal hypoglycemia. Over or under supplementation with sugar would potentially damage the brain [
9,
31]. Therefore, how to balance the risks is still a challenge [
32]. Among our results, there were no significant differences in sex, gestation, birth weight, Apgar score, or cases of breast feeding between the neonatal hypoglycemia group and the normal control group. However, significantly more mothers used insulin during the perinatal period in subgroup A3 than in subgroups A2 and A3 (31% vs. 2%,
p < 0.0001; 31% vs. 7.9%,
p = 0.027, respectively). In addition, the weight gain of the mother during pregnancy in subgroups A2 and A3 was significantly higher than that in the control group (15.3 ± 1.9 kg vs. 11.1 ± 2.2 kg,
p = 0.0154; 14.8 ± 2.6 kg vs. 11.1 ± 2.2 kg,
p = 0.0342, respectively). No obvious difference between subgroup A1 and the control group was found (112.4 ± 3.5 kg vs. 11.1 ± 2.2 kg,
p = 0.7452).
The limitations of this study include a small sample size for the subgroups, the study being conducted in a single center, and a short follow-up time of 2 years. Therefore, observation of the long-term effects of neonatal hypoglycemia is necessary because fine motor development of children occurs until they reach 8 years old.